A DEMOGRAPHIC AND SOCIOECONOMIC DESCRIPTION OF THE WOMEN
IN COLORADO WHO HAVE DIED OF HIV/AIDS
Katherine Ann Lineberger
B.A., Metropolitan State College, 1990
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
1995 by Katherine Ann Lineberger
All rights reserved.
This thesis for the Master of Arts
Katherine Arm Lineberger
has been approved
Lineberger, Katherine Ann (M.A., Sociology)
A Demographic and Socioeconomic Description of the Women
in Colorado Who Have Died of HTV/AIDS
Thesis directed by Assistant Professor Candan Duran-Aydintug
Women are the fastest growing group in the U.S. to be contracting HTV. This fact poses
several problems because of womens secondary access to class, status, and power in this country.
Within the community of women the HIV/AIDS epidemic seems to have concentrated in women of
color and low socioeconomic status; however, little is known about these women as government,
researchers and community agencies only recently have begun to address them. Beginning January
1987, the CDC affirmed the diagnosis HTV/AIDSas a legitimate cause of death category. That same
year (1987) the CDC updated its definition to include some of the disorders women were suffering
from. By changing the definition again in 1993, the CDC made possible the more correct and earlier
diagnoses of women, allowing for more medical intervention throughout the course of the disease.
Now, more than ever, it is possible to obtain a clear, accurate view of this population. In this study
state death certificates are used (since January 1987) of women in Colorado with HIV/AIDS (or
related complications) to identify a detailed description of this population. Descriptive statistics of
age, marital status, residence, race/ethnicity, socioeconomic status, and every possible combination of
these are reported. Among the findings was that the mean age of the population was 41.87, the
median was 39, and the mode was 35, but when age was analyzed by race, these statistics showed a 5-
7 point differential in favor of White women. Also, the majority of women were married (40%), 20%
were single, 11% were widowed, and 29% were divorced. In addition, use of the Hollingshead Two
Factor Index of Social Class showed that these women fell into the middle to working classes, with
24% falling into Class Level H, 34% into Class Level in, aiid 42% into Class Level IV. Despite its
shortcomings (such as the use of incomplete measurement of race/ethnicity on Colorado death
certificates), this study is a pioneer in the field and offers a complete, thorough picture for researchers
and policy makers to understand this population.
This abstract accurately represents the content of the
candidates thesis. I recommend its publication.
This thesis is dedicated to my father, John Lineberger and my mother and stepfather, Mary and John
Chaisson for their unfailing support of and dedication to me during all my academic endeavors.
1. INTRODUCTION AND A STATEMENT OF THE PROBLEM........................1
2. REVIEW OF THE LITERATURE AND CONCEPTUAL FRAMEWORK..................3
Methods Used by Other Researchers................................13
Privacy and Confidentiality......................................16
Race, Ethnicity, and Origin.............................. 18
HTV/AIDS Incidence Per Year, 1987-1992....................20
Manner of Death.......................................... 21
Occupation and Industry...................................21
Common Illnesses in the Population........................21
4. ANALYSES AND RESULTS..............................................22
Race, Ethnicity, and Origin.
Marital Status..................................................... 25
Occupation and Industry..............................................36
Manner of Death.................................................... 37
HIV/AIDS Incidence per Year, 1987-1992...............................37
Common Illnesses in the Population...................................42
Limitations, Justifications, and Conclusions.........................51
A LETTER OF PROTOCOL TO THE COLORADO
DEPARTMENT OF HEALTH...'.............................53
B. SAMPLE COLORADO DEATH CERTIFICATE.......................55
4.3 Race and Origin...................................................................24
4.4 Ages at Death by Race.............................................................26
4.5 Ages at Death by Race and Origin..................................................27
4.6 Marital Status by Race and Origin.................................................31
4.7 Residence at Death by Race and Origin.............................................32
4.8 Years of Education by Race and Origin.............................................35
4.9 Socioeconomic Status: Utilizing the Hollingshead Two Factor Index of Social Class.36
4.12 Manner of Death by Race and Origin................................................40
4.13 HIV/AIDS Incidence per Year, 1987-1992 by Race and Origin.........................41
4.14 Number and Percent of Incidents by Disease Name and Range Codes...................44
Several people deserve acknowledgment with regard to this thesis. First and foremost, I
would like to thank Dr. Candan Duran-Aydintug for taking me under her wing. Without her
friendship and assistance I would surely have floundered much more than I did. Next, the
Department of Sociology at the University of Colorado at Denver deserves kudos for their support, as
I know I can be a difficult student at times. Thanks, also, to M. J. Lane for bringing her computer
wizardry into my life. As well, I would like to thank Stephen O. Murray, for allowing me to utilize
his expertise in editing this paper. Finally, I would sincerely like to thank Joy Breeze for saving my
academic career on two occasions when I was in danger of death by paper work and for sharing
poetry with me when I was exasperated.
INTRODUCTION AND A STATEMENT OF THE PROBLEM
There is very little information available on women in the United States with Human
Immunodeficiency Virus (HIV), the virus which causes AIDS (Acquired Immune Deficiency
Syndrome), and the state of Colorado is no exception. The Colorado Department of Health publishes
a quarterly newsletter entitled AIDS: Status in Colorado, which lists the number of reported cases
of HTV/AIDS by sex and race, the current mortality rate, transmission categories, and geographic
regions. While this information is helpful, it is scant in its description of the women in Colorado
with HTV/AIDS. Clearly, a more detailed and accurate description must be made for two reasons.
First, if any further research into this area is to be done, researchers should be familiar with their
population. At present, this would be nearly impossible because of the lack of information about
women with HTV. Second, if any policy is to be instituted with regard to this population of people,
policy makers will need concrete evidence about who is to qualify for such policy.
There is a serious flaw in the methodology used to compile statistical information about
women with HTV/AIDS in the United States. Statistical data are generated using reported numbers.
This means that, whenever a person is tested for HTV and the results are positive, that information is
reported to the Centers for Disease Control (CDC). There is a law in Colorado which states that any
person found by a physician to have HTV or AIDS must be reported to the Department of Health.
When a case is reported, the Colorado Department of Health then contacts the infected person and
requests their sexual history, along with names, addresses and phone numbers of each of their sexual
partners, since having contracted HIV. Gena Corea describes in her book The Story of Women and
AIDS: The Invisible Epidemic, that many women are not tested for HTV or identified as having
AIDS because they do not fit the description of a high risk person (Corea, 1992). Also, many people
choose against seeking diagnosis or treatment when they suspect that they have HTV/AIDS because
they want to forego the intrusion of the Department of Health into their lives. Furthermore, many
doctors will not report the positive test results of their patients because they morally disagree with
mandatory reporting. It is for these reasons that any reported statistics the Colorado Department of
Health are more likely to be not quite accurate.
While the information available from the Department of Health is valuable, it is, for the most
part, non-gendered or male-centered information. This leads to several questions. What races
ethnicities are the women in Colorado with HIV/AIDS? How old are these women? What is their
marital status? Where do they live? What opportunistic infections are the most common ones for
women in Colorado? What is the current mortality rate for women with HTV/AIDS in Colorado?
Lastly, how is the variable of socioeconomic status displayed among these women? These are the
questions that this study addresses.
REVIEW OF THE LITERATURE AND CONCEPTUAL FRAMEWORK
Women are the fastest growing segment of the population in the United States to be
contracting HIV/AIDS. This poses several problems because of womens secondary access to class,
status, and power in this country. These concepts are important to note, as well, because within the
community of women, the HIV/AIDS epidemic seems to be concentrated in women of color and of
low socioeconomic status. Little is known about these women as government, researchers, and
community agencies have only recently begun to address them. According to Gena Corea in her work
The Invisible Epidemic: The Story of Women and AIDS, dining the 1980s, there was a tremendous
increase in the number of women complaining of chronic vaginal and pelvic pain. However, most of
these women were not tested for HIV because they did not fit the description, set forth by the Centers
for Disease Control (CDC), of a high risk group. Some survivors have even told of requesting that
their doctors perform an HTV antibody test on them, only to be condescendingly refused, as if they
did not know what they were talking about. Some women were told that their pain was
psychosomatic and that they should seek psychotherapy to cure their mysterious malady (Corea,
In Dangerous to Your Health: Capitalism in Health Care. Vicente Navarro discusses the fact
that the United States uses race to classify mortality and sickness rates, but rarely includes analysis of
class. He claims that class is a very important indicator of the health status of an individual in this
country and that the reason it is ignored in issues of prominence is because the higher classes have
more power than the lower and suppress this information (Navarro, 1993). S. Leonard-Syme and
Lisa F. Berkman, who wrote the review Social Class, Susceptibility, and Sickness, discuss the
connection between social class and mortality and morbidity. They cite a nationwide study tty
Kitawaga and Hauser (1973) in which they found mortality rates to be extremely differential between
and among people of different ages, occupations, levels of education, and income. Leonard-Syme
and Berkman (1994) conclude that lower class groups do have higher morbidity and mortality rates of
almost every disease and illness, and that this stands true over time. Lastly, Irving Kenneth Zola, in
Medicine as an Institution of Social Control, discusses how medicine has traditionally held a de
facto relationship with social control. In this relationship, members of the medical professions
control who they treat with what and what they choose to treat (Zola, 1994). This body of
information highlights implications for the issue of women and HrV/AIDS,.the lack of attention and
monies being invested in researching and treating women, and the fact that the mortality rate for
women in this population is rising so quickly.
In the book A Sociology of Women: The Intersection of Patriarchy. Capitalism, and
Colonization. Jane Ollenburger and Helen Moore discuss how women hold the low status positions in
employment. Women continue to make approximately seventy-six cents to a mans dollar, even
though this disparity was discovered years ago. As a result, women hold a very low class level in the
United States and this is especially true for women of color. Also, because they do not contribute as
much to the economy, womens needs are seen as secondary to those of the white males who hold
significant financial status. It is in the interest of the upper/male class to keep the lower/female class
in subjugation, for without it, they would lose their place at the head of the line. It is also in mens
interest to keep women in the lower strata because patriarchy is beneficial for men. Women do the
bulk of family work (e.g. housecleaning, child rearing, and entertainment), which frees men to pursue
their careers and activities of choice. Women also manage to occupy the positions in the workplace
which men consider to be of little import and this phenomenon also leaves men free to pursue their
desires (Ollenburger and Moore, 1992).
In the lives of women of color, this disparity increases drastically. While white, middle class
women tend to hold the more professionalized employment positions of the lower strata, such as
clerical work, women of color hold more manual labor positions, such as housecleaning, which pay
even less than clerical positions. In addition, women, in general, and women of color in particular,
are increasingly becoming the heads of households with children. At least fifty percent of woman
headed households have incomes below poverty level (Harris, 1993) and child support payments by
fathers are often inconsistent or nonexistent (Kitson, 1992; Ollenburger and Moore, 1992). These
reasons, coupled with unemployment or underemployment, lead many women to take work anywhere
they can get it. Often times, this work may mean the streets, explains ACT UP New York, which
published Women. ADDS, and Activism (1990). The combination of these frequently leads women to
abuse drugs which temporarily dull the pain of their undervalued lives and the lives that they provide
to their children (ACT UP New York, 1990). Unfortunately, these are the faces of too many women
with AIDS. Because of their low socioeconomic status and the unequal race distribution of
HIV/AIDS, their health care needs go uncared for and unchecked, except in where they coincide with
the health care of those in power.
HIV/AIDS seems definitely to be a class issue and women, as a segment of the population,
are devalued in relation to men in the United States. Initially, the disease was practically ignored
because it was discovered in the gay male community, a community which holds a status below that
of heterosexual males. Then, the disease was seen in more innocent communities, such as blood
and plasma recipients. When this happened in the mid to late 1980s, more attention, time, and
money was paid to the disease by the government. At the present time, when HTV is again growing
fastest in a lower class of people, namely women, less attention and fewer monies are being invested
in research, education, and public policy. It seems health care is, in fact, by the rich, for the rich, and
of the rich.
A prominent issue with regard to gender and health care is that women, especially lower
income minority women, do not have the access to health care that men do. Many times, women
without health insurance depend on state operated hospitals to care for them and often suffer lower
standards of care as a result. Poor women, because they cannot afford the often high cost of medical
procedures for their maladies, are frequently not even offered treatment that could save their lives.
Because women in the United States with HIV/AIDS are concentrated in the lower strata, they must
deal with having a debilitating and life threatening disease and have little to no hope for treatment of
their disease related conditions.
Women, until recently, were completely ignored in the Centers for Disease Control (CDC)
definitions of HIV/AIDS. Since the disease was first thought to be concentrated only in the gay and
bisexual male community, the CDC collected information and reported about only the manifestations
of the disease in this particular community. What is significant about this is that men tend to contract
AIDS related diseases such as Kaposis Sarcoma (KS) and Pneumosystis Carinii Pneumonia (PCP),
whereas women tend to contract gynecological manifestations of AIDS, such as Cervical Cancer and
Pelvic Inflammatory Disease (PID).
A typical attitude of the health care sector toward women is that of hysteria. Women are
thought to be out of their minds if their illness does not make sense to a male doctor (Ehrenreich and
English, 1973). In the beginning of the HIV epidemic, women who went to their doctors with
unexplained and chronic vaginal infections were often told that their symptomology stemmed from
their minds, to seek a therapist, or that perhaps it was an unconscious resentment of their femaleness.
Later, when women were recognized as carriers of HTV, treatment for their related conditions was
often refused by doctors. Treatment was seen as unproductive, as the ill women were invariably to die
anyway. So, women died, not only earlier than necessary, but in tremendous pain from AIDS related
complications that could have been treated (ACT UP New York, 1990).
As some doctors and activists in the womens community began to suspect that women
might be contracting HTV, they sought help from the Centers for Disease Control (CDC) to do
research on women. They were told that, since women did not fit the description of high risk people,
there was no reason to research them with regard to this disease (Corea, 1992). Even as women
became increasingly identified as having the disease, the CDC stated that, because they were not
dying from illnesses such as Kaposis Sarcoma (KS), they were not officially dying from AIDS. In
fact, by 1990 only about six and one-half percent of the clinical trial medical research done on
HTV/AIDS was performed with woman subjects. Of these, approximately ninety percent were studies
focused on the manifestation of HTV/AIDS in pregnant womens fetuses, rather than on the
manifestations of the disease in the women themselves (Corea, 1992; Zarembka and Franke, 1990).
Nevertheless, doctors and researchers were asserting that they were definitely observing a different
pattern in women, and the CDC continued to ignore them (Corea, 1992).
In the midst of all the arguments women were increasingly becoming infected with the HIV
virus. Usually, when they became too debilitated to work, they applied for governmental social
service assistance and were refused, again because they did not fit the official description of someone
with HIV/ADDS. They sought help from community programs that serviced People With AIDS
(PWAs) and again were turned down for the same reason and because no money was being allocated
for the treatment and care of women and the disease (Corea, 1992). If a woman did not have a strong
family or friendship unit to fall back on, she often found herself homeless, with no hope for help with
her fatal disease. Also, a final, devastating blow to women was that their children were often taken
from them by the state when they became unable to care for them any longer. Increasingly, women
with AIDS found themselves completely rejected by society.
Ironically, whenever women have been addressed with regard to HTV/AIDS issues, they have
been perceived as vectors of the disease. An interesting reaction of our society towards persons with
AIDS has been to place blame on the sufferers not only for the effects of the disease on themselves,
but also the effects of the disease on society as a whole. Weiss and Lonnquist (1994) state that,
because HTV/AIDS has primarily affected people in society who are considered deviant, blame for the
disease is, in fact, very often put on the victims themselves. This dichotomy of the guilty and the
innocent diseased came about as a result of social stereotypes, fears, and political biases which are
in reaction to seeming threats of political or social change. It could be said that the United States
response to HIV is one of the most accurate mirrors of our collective prejudiced psyche. Bowleg
(1992) discusses how female prostitutes have been condemned for passing HIV on to unsuspecting
customers, without even a thought for the customers role in the exchange or to how the prostitute,
herself, got it. The fact which makes this assertion so outrageous is that HTV passes at least seventeen
times more efficiently and effectively from a male to a female, rather than from a female to a male
(ACT UP New York, 1990). Also, according to Bowleg (1992), most prostitutes in the United States
are poor women of color, a fact which brings credence to assertions that poor women of color are
being discriminated against in the epidemic.
The literature that has been produced to date on women and HTV/AIDS is very sketchy, but
helpful. It is not particularly sociological in nature but it does provide a good basis for such research.
Unfortunately, the existing empirical research in this area is quite deficient. In fact, almost all of the
research discussions of this subject mention a lack of detailed description and the need for further
research (Gayle, et al, 1991; Murphy, 1988). Hankins and Handley describe the empirical literature
as case reports, cross-sectional studies, and retrospective evaluations published primarily in
conference proceeding with no detailed descriptions of research methodology (1992). Outside of the
biomedical research, which in itself is sparse, the psychosocial methodology has mostly been
concentrated in the area of survey, and the bulk of this has been focused on the sexual politics of
condom use. Another area of focus has been that of women and intravenous drug use (TVDU).
However, as can be seen, the majority of the literature is largely inclusive of opinion papers, mixed in
with some theory, and little that is empirical in nature.
Among the empirical research that has been published is a paper entitled Women and HIV
Disease, written by Dr. Barbara Cambridge. Dr. Cambridge has been leading support groups for
women with HTV/AIDS since 1987 and designed the paper out of her participant research of these
women. The paper includes a brief, yet thorough, overview of the issues of women with HIV. She
states that among these issues are health care and personal wellness. In this section, Cambridge
describes how women with AIDS often have a difficult time connecting with the health care system
in the United States, which leads to a quicker and more painful disease progression for women. She
also mentions some of the emotional challenges facing women with HTV, including fear of dying and
of abandonment, loss of sexual intimacy, and reproductive issues. Finally, she discusses how
unemployed and underemployed women do not have access to health insurance and often cannot
afford treatment for their HTV related conditions. Cambridges paper is directed toward caregivers of
women with HTV/AIDS. It is clear, concise, and brief; similar to Gena Coreas work but without the
in depth history. Unfortunately, when compared to Coreas, Cambridges work may seem redundant.
Another unfortunate aspect of this paper is that it is very weak on methodology. Many questions are
left unanswered in the reading. For instance, how did Cambridge fill her therapy groups? Were they
referred patients? If yes, referred by whom? Did she seek out women with HTV? If yes, where?
Were there specific differences between race, socioeconomic status, or geographical region? and so
on. Of course, these questions can be answered with further empirical research.
In another research project pertinent to this study is Surveillance for AIDS and HIV
Infection Among Black and Hispanic Children and Women of Childbearing Age, 1981-1989. Here,
Gayle, et al surveyed HTV/AIDS data from four sources:
1. National AIDS surveillance data reported to the division of HTV/AIDS, Center for
Infectious Diseases, CDC, in 1989.
2. Mortality data reported to the National Center for Health Statistics, CDC, in 1988.
3. Data on HIV-prevalence in specimens routinely submitted for metabolic screening
of newborns from December 1987 through November 1988.
4. Data on HTV-antibody prevalence of women who applied for military service from
October 1985 through December 1989.
Analysis of the data found disparity in the mortality rates between African-American women
and white women, across ages 15 to 24, 25 to 34, and 35 to 44, in favor of white women. In all, the
researchers estimated that HTV-related deaths occurred for eleven percent and three percent of all
deaths among black and white women, respectively. In conclusion, they state that their study is a
good indication of the need for prevention strategies targeted to different racial, age, and ethnic
Gayle, et als study was contributive in identifying and describing some of the differential
variables among and between women with HIV/AIDS. In addition, it was a creative use of secondary
data sources. Furthermore, it again confirms that women of low socioeconomic status and women of
color may be at higher risk for HTV than are white, middle class women. However, it just begins to
entice with this information before completion. Their sampling procedures leave much to be desired.
First, it is already known that the National AIDS surveillance data are questionable. Little from the
Centers for Disease Control, the Department of Health, or the Department for Health Statistics can be
held as a complete and reliable description of the population of people with AIDS because of issues
related to reporting positive cases. Second, using the antibody prevalence of women applying to the
military seems very limiting and has little external validity. There seems to be no justification for the
use of such data, and its significance for women with HIV in general is questionable. It would seem
that women applying for military service would be quite a specific group and that any information
about this group would apply to this group only, and not to any larger, more general population.
What is lacking in almost all of the literature and research on HIV in women is an accurate,
quantitative and qualitative description of this population. The methods used to date in attempts to do
this are very problematic. In this study, an attempt is made to deal with these problems.
Methods Used bv Other Researchers
Ironically, it was after a methodology for this project had been decided upon that a similar
study was discovered. The American Journal of Public Health published a study by James W.
Buehler, et al., in which Buehler and his associates utilized vital statistics to assess the accuracy of
the reporting of HIV/AIDS in women. They used death certificates and the reports of cases of
HTV/AIDS in women from the Centers for Disease Control (CDC), searching for one of three types:
1. Women with HTV/AIDS listed as the primary cause of death. .
2. Women with HTV/AIDS listed as a secondary complication of death.
3. Women who died of ailments common to women with HTV/AIDS (e.g. Cervical Cancer).
For the year 1988, for women ages 15 through 44, there were 1,365 deaths attributable to
HTV/AIDS as a primaiy cause, 202 deaths with HTV/AIDS listed as the secondary complication, and
149 excess due to conditions common to women with HTV/AIDS. Of the deaths that occurred in
1988, 1,532 were reported to the CDC (1,668 when adjusted for report delays). In conclusion, the
researchers state that underlying cause of death vital records and AIDS surveillance identified fifty-
five to eighty percent and sixty-seven to ninety-seven percent, respectively, of HTV/AIDS related
deaths in women, ages 15 through 44. The wide ranges of these estimates reflect the potential role of
HTV infection in contributing to excess mortality (1992).
While this information is valuable and affirms the fact that women are the fastest growing
population in the United States to be contracting HIV/AIDS, it nevertheless has some problems.
Gena Corea explains that many doctors do not even think to test their female patients for HIV. Many
people in the health professions continue to hold on to the misinformed idea that people with AIDS
come from high risk groups rather than people who engage in high risk behavior (1992). The
difference in reported rates of the first, misinformed definition and the latter, more accurate definition
could be quite considerable. As a result, there is reason to believe that many women who do have the
HTV virus will die without having been thoroughly and accurately diagnosed during their illness. It
is also because of this that the Buehler, et al. study may not be totally representative of the nations
women with HIV/AIDS. A second problem with this research is that it is limited in scope. While the
researchers were investigating, it might have been simple to compile descriptive information about
these women, yet they have not. The data were available for them to record the race, occupation, and
more of each of the recorded deaths but they chose not to access this information. The present study
addresses these additional variables. Also, similar to the Buehler, et al. study, this study utilizes vital
statistics, in the form of death certificates, to perform this descriptive analyses.
Beginning January 1987, the Centers for Disease Control (CDC) affirmed the diagnosis
AIDS or HTV as a legitimate cause of death category, to be used on death certificates. Since that
time, the CDC has twice updated its definition of HTV/AIDS, once in 1987 and again in 1993. In
1987, the changes in definition included the addition of some of the ailments observed in women
infected with HIV. In the past, the CDC identified HTV/AIDS related disorders most common to
men, since the epidemic was discovered in the gay male community. What researchers concerned
with womens health suspected is that women manifest HTV/AIDS in different way than men,
becoming more often ill with gynecologically related illnesses. However, even with the change in
definition, the power of it to identify the prevalence of HTV/AIDS in women continued, largely, to be
weak. In spite of this, the allowance of HTV/AIDS as a cause of death category on death certificates
made it possible to get a much more clear and accurate view of the population of women in the United
States who have HIV/AIDS.
The use of state death certificates to determine a clear and concise description of the women
in Colorado with HIV/AIDS surpasses the use of reported data on this subject. The state of Colorado
holds death certificates as confidential, yet the Department of Health has the computer capacity to
access aggregate data about deaths in the state, thereby assuring individuals privacy. It is also
assumed that peoples reservations about privacy issues, with regard to reporting HTV/AIDS positive
cases, would be relayed once death has occurred. Because of this, some of the gaps in the reported
statistics were filled, ex post facto, in death certificate reports. Additionally, Colorado death
certificates include voluminous information about the deceased, enabling the determination of
demographic, geographic, and socioeconomic information about the target population. Again,
because of the many problems associated with reported statistics (e.g. non-reporting and
misinformation), this information should be more accessible and trust worthy in death certificate
As previously stated, the Colorado Department of Health has the computer capacity to access
aggregate data collected from all death certificates in the state, which bypasses the need for sampling.
Consequently, the population is defined as all female in Colorado from January 1,1987, through
December 31, 1992, who died:
1. With HIV/AIDS noted as the secondary cause of death.
2. With HTV/AIDS noted as the underlying cause of death.
3. With HIV/AIDS noted as an aggravating condition present at the time of death.
4. With HIV/AIDS noted as having been discovered upon autopsy.
5. With HTV/AIDS noted in any place on the death certificate.
Privacy and Confidentiality
Two procedures were followed in order to comply with Colorado law governing the
confidentiality of death certificate information. First, a letter of protocol was sent to:
The Colorado Department of Health
Health Statistics Section
4300 Cherry Creek Drive South
Denver, Colorado 80222-1530
(See Appendix A for a copy of this letter). Once the letter and proposal were sent, reviewed, and
accepted, a letter of confidentiality was signed before the data were released and research was begun.
See Appendix B for a sample copy of a Colorado Death Certificate. The following variables
are available through the Colorado Department of Health for analysis:
The Colorado Department of Health indicates the exact age, in years, of the deceased on the
death certificate form, enabling measurement and analysis of this variable.
Death certificates from the Colorado Department of Health operationalize marital status as
Marital Status Unknown
The Colorado Department codifies residence according to Metropolitan Statistical Area
(MSA). This information is available in the following form:
1. Denver Primary Statistical Area: This includes the five counties of Adams, Arapahoe,
Douglas, Jefferson, and Denver.
2. Denver Consolidated Statistical Area: This includes the six counties of Adams,
Arapahoe, Douglas, Jefferson, Denver, and Boulder.
3. Weld County Metropolitan Statistical Area.
4. Larimer County Metropolitan Statistical Area.
5. El Paso County Metropolitan Statistical Area.
6. Pueblo County Metropolitan Statistical Area.
7. Total Metropolitan Area.
8. Total Non-Metropolitan Area.
Race. Ethnicity, and Origin
The Colorado Department of Health operationalizes race and ethnicity as follows:
1. White (includes Mexican, Puerto Rican, and all other Caucasian).
2. Black, Negro, Colored, Afro-American.
3. Indian (American, Alaskan, Canadian, Mexican, Eskimo, and Aleut).
6. Hawaiian (includes Part-Hawaiian).
7. Other Nonwhite (includes Cajun, Creole).
9. Unknown, not stated, or not classifiable.
10. Other Asian or Pacific Islanders (Korean, Thai, Vietnamese, Chamorro, Guamian, etc.).
The Colorado Department of Health operationalizes origin as follows:
2. Puerto Rican.
4. Central and South American.
5. Other and Unknown Spanish.
7. Indian (American, Alaskan, Canadian, Mexican Indian, Eskimo, and Aleut).
8. British, Scottish, Welch, and Scotch-Irish.
12. Norwegian, Swedish, and Danish.
15. Other North, Central, and South American.
16. Other Western European.
17. Other Northern European.
18. Other Eastern European.
19. Other Southern European (excludes Spain).
20. Southeast Asian and Pacific Islander.
21. South Central Asian.
22. Other Asian.
23. North African.
24. Other African.
27. Unknown. /
Socioeconomic Status /
The Colorado Department of Health lists the total number of years of education the decedent
attained, the decedents usual occupation, and the type oljbusiness or industry that the decedents
usual occupation is categorized as. The Hollingshead Two Factor Index of Social Class was used in
conjunction with these data to determine a measure of socioeconomic status.
HIV Incidence Per Year, 1987-1992 j
Because the data from the Colorado Departmentjof Health reflects the number of deaths per
year, from 1987 through 1992, fluctuations and basic trends are readily observable.
The Colorado Department of Health notes the total number of years of education attained by
the decedent, if possible.
Manner of Death
The Colorado Department of Health operationalizes the manner of death variable as follows:
5. Pending Investigation.
6. Undetermined Manner.
Occupation and Industry
The Colorado Department of Health has constructed a code booklet for occupation which is
based on the 1980 Standard Occupational Classification Manual. Also, to analyze industry, they have
compiled a code booklet based upon the 1987 Standard Industrial Classification Manual. This
enables the discovery of which occupations and industries are most common to women in Colorado
Common Illnesses in the Population
The Colorado Department of Health utilizes the Eighth and Ninth Revisions of The
International Classification of Diseases to codify the primary, secondary, and underlying causes of
death on death certificate forms. In addition, they utilize this same coding for aggravating conditions
known at the time of death. This enables the study of which opportunistic infections are common to
women in Colorado with HIV/AIDS.
ANALYSES AND RESULTS
Decoding of the data was simply a matter of utilizing the code books which the Colorado
Department of Health uses in their data analyses. There were a total of 75 death certificates to
analyze and all were accomplished using the Microsoft Excel Five Program. Frequency distributions
were organized for every variable and every possible combination of variables. Also, the mean,
median, and mode were calculated for every possible variable and the combinations. Following
calculation, the findings were put into table form.
Race. Ethnicity, and Origin
The Colorado Department of Health codes the variable of race using The Data Management
Section Statistical Abstract File Description. 1970 Throueh the Current Year (The Colorado
Department of Health, 1989). For a total of seventy-five women, fifty-four (72%) were listed as
White and twenty-one (28%) as Black (see Table 4.1). Regarding origin (see Table 4.2), fifty-one
women (68%) were described as Nonhispanic, twelve (16%) as American, two (3%) as Mexican, one
(1%) as Irish, eight (11%) as Other and Unknown Spanish, and one (1%) as Other Southern
European (excludes Spain). Of course, when the two variables were combined, the results were more
illuminating. One (1%) woman was listed as White/Irish, one (1%) was listed as Black/Other
Southern European (excludes Spain), two (3%) were listed as White/Mexican, four (5%) as
Black/Other and Unknown Spanish, four (5%) as White/ Other and Unknown Spanish, sixteen
(21%) as Black/Nonhispanic, twelve (16%) as White/American, and thirty-five (47%) as
White/Nonhispanic (see Table 4.3).
Table 4.1 RACE
WHITE 54 72%
BLACK 21 28%
TOTAL 75 100%
Table 4.2 ORIGIN
NONHISPANIC 51 68%
AMERICAN 12 16%
OTHER & UNKNOWN 8 11%
MEXICAN 2 3%
IRISH 1 1%
OTHER SOUTHERN 1 1%
EUROPEAN (Excludes Spain)
TOTAL 75 100%
RACE AND ORIGIN
WHITE / NONHISPANIC 35 47%
WHITE /AMERICAN 12 16%
BLACK / NONHISPANIC 12 16%
BLACK/OTHER & UNKNOWN SPANISH 4 5%
WHITE / OTHER & UNKNOWN SPANISH 4 5%
WHITE/MEXICAN 2 3%
BLACK / OTHER SOUTHERN EUROPEAN 1 1%
WHITE /IRISH 1 1%
TOTAL 75 100%
The Colorado Department of Health codes the variable of age using The Data Management
Section Statistical Abstract File Description. 1970 Through the Current Year (Colorado Department
of Health, 1989). All seventy-five death certificates have the age of the decedent recorded. The
mean and median age for the total population were both 69.5. Interestingly, these figures changed
when the variable was analyzed by race. White women had a mean and median age of 57. However,
black women had a mean and median age of 50.5.(see Table 4.4). And when the decedents origin
was added to the calculation, the results showed that White/Nonhispanic women had a median age of
33, White/American women had a median age of 54, White/Other and Unknown Spanish women had
a median age of 50, White/Mexican women had a median age of 39.5, and there was only one
White/Irish woman whose age was 69 (see Table 4.5).
The Colorado Department of Health codes the variable of marital status using The Data
Management Section Statistical Abstract File Description. 1970 Through the Current Year (The
Colorado Department of Health, 1989). Marital status was recorded on seventy-five of the records, as
well. Thirty of the women were married, which equaled 40% of the population. Fifteen women
(20%) were single, eight women (11%) were widowed and, finally, 22 women (29%) were divorced.
Of the married women, eleven were White/Nonhispanic, seven were Black/Nonhispanic, six were
White/American, two were White/Other and Unknown Spanish, and White/Irish, Black/Other
Southern European, and Black/Other and Unknown Spanish had each one incident. Of the single
women, ten were White/Other and Unknown Spanish, two were White/Americans, two were
Black/Nonhispanics, and two were Black/Other and Unknown Spanish (see Table 4.6).
AGES AT DEATH BY RACE
AGE WHITE BLACK TOTAL
<1 1 0 1
1-4 1 0 1
5-9 0 0 0
10-14 1 0 1
15-19 0 0 0
20-24 2 0 2
25-34 11 8 19
35-44 15 8 23
45-54 11 3 14
55-64 4 1 5
65-74 7 1 8
75-84 1 0 1
85+ 0 0 0
TOTAL 54 21 75
MEDIAN 57 50.5
AGES AT DEATH BY RACE AND ORIGIN
AGE RAN GE WHITE/NON HISPANIC WHITE/ AMERICAN WHITE/ OTHER& UNKNOWN SPANISH WHITE/ MEXICAN SUBTOTAL
Num ber Per cent Num ber Per cent Num ber Per cent Num ber Per cent Num ber Per cent
<1 1 1% 0 0% 0 0% 0 0% 1 1%
1-4 1 1% 0 0% 0 0% 0 0% 1 1%
5-9 0 0% 0 0% 0 0% 0 0% 0 0%
10- 14 1 1% 0 0% 0 0% 0 0% 1 1%
15- 19 0 0% 0 0% 0 0% 0 0% 0 0%
20- 24 2 3% 0 0% 0 0% 0 0% 2 3%
25- 34 6 8% 4 5% 0 0% 1 1% 11 13%
35- 44 10 13% 4 5% 1 1% 0 0% 15 20%
45- 54 8 11% 0 0% 2 3% 1 1% 11 13%
55- 64 3 4% 0 0% 1 1% 0 0% 4 5%
65- 74 3 4% 3 4% 0 0% 0 0% 6 8%
75- 84 0 0% 1 1% 0 0% 0 0% 1 1%
85+ 0 0% 0 0% 0 0% 0 0% 0 0%
TOT AL 35 47% 12 16% 4 5% 2 3% 53 71%
MED IAN 33 54 50 39.5
Table 4.5 AGE AT DEATH BY RACE AND ORIGIN (Continued)
AGE RAN GE WHITE/ IRISH BLACK/ NON HISPANIC BLACK/ OTHER& UNKNOWN SPANISH BLACK/ OTHER SOUTHERN EUROPEAN (excludes Spain) SUBTOTAL TOTAL
Num ber Per cent Num ber Per cent Num ber Per cent Num ber Per cent Num ber Per cent Num ber Per cent
<1 0 0% 0 0% 0 0% 0 0% 0 0% 1 1%
1-4 0 0% 0 0% 0 0% 0 0% 0 0% 1 1%
5-9 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
10- 14 0 0% 0 0% 0 0% 0 0% 0 0% 1 1%
15- 19 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
20- 24 0 0% 0 0% 0 0% 0 0% 0 0% 2 3%
25- 34 0 0% 6 8% 2 3% 0 0% 9 12% 20 27%
35- 44 0 0% 7 9% 1 1% 0 0% 8 11% 23 31%
45- 54 0 0% 2 3% 1 1% 0 0% 3 4% 14 19%
55- 64 0 0% 0 0% 0 0% 1 1% 1 1% 5 7%
65- 74 1 1% 1 1% 0 0% 0 0% 2 3% 8 11%
75- 84 0 0% 0 0% 0 0% 0 0% 0 0% 1 1%
85+ 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
TOT AL 1 1% 17 21% 4 5% 1 1% 23 31% 75 100
MED IAN 69 41.5 62
The Colorado Department of Health codes residence using Vital Statistics Countv / City
Table (The Colorado Department of Health, 1988). A total of sixty-two cases were analyzed for the
general residence variable. Twenty-nine of the cases were from Denver County, twelve were from El
Paso County, nine were from Jefferson County, three were from Weld County, and two each were
from Larimer and Garfield Counties. Douglas, Fremont Canon, Mesa, Moffa, Montezuma, and
Sagua Counties each listed one death. As with age, this variable led to different results whenever
analyzed with race. For a total of 64 cases analyzed as residence by race, eighteen cases were Black
women and forty-six were White women. Of the Black women, fourteen (2.2% of the total
population) resided in Denver and four (5% of the total population) resided in El Paso county.
Denver, El Paso, and Jefferson Counties had the largest number of White women in the study, with a
total of thirty-one (41% of the total population). Of the White women, fifteen (20% of the total
population) resided in Denver County, seven (9% of the total population) resided in El Paso County,
and nine (12% of the total population) resided in Jefferson County. When analyzed by race and
origin, Denver County had one White/Irish woman, three White/American women, eleven
White/Nonhispanic women, twelve Black/Nonhispanic women, one Black/Other and Unknown
Spanish woman, and one Black/Other Southern European (excludes Spain). El Paso County had six
White/Nonhispanic women, two Black/Nonhispanic women, one White/American woman, and one
Black/Other and Unknown Spanish woman. Jefferson County had one White/American woman, one
White/ Other and Unknown Spanish woman, and seven White/Nonhispanic women. Weld County
had one White/American woman and two White/Nonhispanic women. Garfield County had one
White/American woman and one White/Nonhispanic woman. Larimer County had two
White/Nonhispanic women. Moffa and Mesa Counties each had one White/American woman.
Fremont Canon and Douglas Counties each had one White/Nonhispanic woman. Sagua and
Montezuma Counties each had one White/Other and Unknown Spanish (see Table 4.7).
The variable of education is codified by the Department of Health using The Data
Management Section Statistical Abstract File Description. 1970 Through the Current Year (The
Colorado Department of Health, 1989). For a total of fifty-nine cases, the mean and median levels of
education of these women were each 8.5. Forty-one (55% of the total population) had high school
level or less education, nine (12% of the total population) had some college or vocational school
training, and nine (12% of the total population) had four years of college education or more. Of the
women who had twelve or fewer years of education, twenty-four (32% of the total population) had a
high school education, six (8% of the total population) had eleven years, five (7% of the total
population) had ten years, four (5% of the total population) had seven to nine years, and women with
five and zero years of education totaled one case each (1% of the total population each). Again, when
analyzed by race, the education variable yields to a sharper image. Of the fifty-nine women for which
these variables could be analyzed, eighteen (24% of the total population) were Black and forty-one
(55% of the total population) were White. The mean and median levels of education for Black
women was 12.5 and,for White women, measured 8.5. Of the Black women, two (3% of the total
population) had four years of college or more, three (4% of the total population) received some
college or vocational training, and fourteen (19% of the total population) had high school or less
education. Of the White women, seven (9% of the total population) had four or more years of college,
six (8% of the total population) had some college or vocational training, and twenty-seven (36% of
the total population) had a high school or less level of education. When combined with race and
origin, the mean level of education for White/Nonhispanic women was 8.5; for White/Other and
Unknown Spanish women was 10; and for White/Mexican women was 12. The mean level of
education for Black/Nonhispanic women was 12.5; and for Black/Other and Unknown Spanish was
11 (see Table 4.8).
Table 4.6 MARITAL STATUS (MS) BY RACE AND ORIGIN
Race MARRIED SINGLE DIVORCED WIDOWED TOTAL
# % # % # % # % # %
W/Nh 11 15% 10 13% 12 16% 2 3% 35 47%
W/A 6 8% 2 3% 3 4% 1 1% 12 16%
w/os 2 3% 0 0% 1 1% 1 1% 4 5%
W/M 0 0% 0 0% 1 1% 1 1% 2 3%
W/I 1 1% 0 0% 0 0% 0 0% 1 1%
B/Nh 7 9% 2 3% 5 7% 2 3% 16 21%
B/OS 1 1% 2 3% 0 0% 1 1% 4 5%
B/ose 1 1% 0 0% 0 0% 0 0% 1 1%
TOT 29 39% 16 21% 22 29% 8 11% 75 100
Legend: W/Nh = White/Nonhispanic W/A = White/American #=Number
W/OS = White/Other & Un- W/M = White/Mexican %=Percent
known Spanish W/I = White/Irish
B/Nh = Black/Nonhispanic B/OS = Black/Other & Unknown Spanish
B/ose = Black/Other Southern European (excludes Spain)
RESIDENCE AT DEATH BY RACE AND ORIGIN
White/ Non- Hispanic White/ American White/ Other & Unknown Spanish White/ Mexican SUBTOTAL
Coun ty # % # % # % # % # %
Den- ver 11 15% 3 4% 0 0% 0 0% 15 20%
El Paso 6 8% 1 1% 0 0% 0 0% 7 9%
Jeffer son 7 9% 1 1% 1 1% 0 0% 9 12%
Weld 2 3% 1 1% 0 0% 0 0% 3 4%
Gar- field 1 1% 1 1% 0 0% 0 0% 2 3%
Lari- mer 2 3% 0 0% 0 0% 0 0% 2 3%
Sa- gua 0 0% 0 0% 1 1% 0 0% 1 1%
Doug las 1 1% 0 0% 0 0% 0 0% 1 1%
Mon- te- zuma 0 0% 0 0% 1 1% 0 0% 1 1%
Mof- fa 0 0% 1 1% 0 0% 0 0% 1 1%
Mesa 0 0% 1 1% 0 0% 0 0% 1 1%
Fre- mont can- on 1 1% 0 0% 0 0% 0 0% 1 1%
TOT AL 31 41% 9 12% 3 4% 0 0% 44 59%
liegend: #=Number %=Percent of 75
RESIDENCE AT DEATH BY RACE AND ORIGIN (Continued)
Black/ Nonhispanic Black/ Other & Unknown Spanish Black/ Other Southern European (excludes Spain) White/ Irish SUBTOTAL TOTAL
Coun ty # % # % # % # % # % # %
Den- ver 12 16% 1 1% 1 1% 1 1% 15 20% 30 40%
El Paso 2 3% 1 1% 0 0% 0 0% 3 4% 10 13%
Jef- fer- son 0 0% 0 0% 0 0% 0 0% 0 0% 9 12%
Weld 0 0% 0 0% 0 0% 0 0% 0 0% 3 4%
Mof- fa 0 0% 0 0% 0 0% 0 0% 0 0% 2 3%
Gar- field 0 0% 0 0% 0 0% 0 0% 0 0% 2 3%
Mesa 0 0% 0 0% 0 0% 0 0% 0 0% 1 1%
Fre- mont Can- on 0 0% 0 0% 0 0% 0 0% 0 0% 1 1%
Lar- imer 0 0% 0 0% 0 0% 0 0% 0 0% 1 1%
Sa- gua 0 0% 0 0% 0 0% 0 0% 0 0% 1 1%
Doug las 0 0% 0 0% 0 0% 0 0% 0 0% 1 1%
Mont ezum a 0 0% 0 0% 0 0% 0 0% 0 0% 1 1%
TOT AL 14 19% 2 3% 1 1% 1 1% 18 24% 62 83%
The Hollingshead Two Factor Index of Social Class was used to analyze the variable of
socioeconomic status. The Colorado Department of Health has occupation and years of education of
the decedent recorded on the death certificate form. Also, the Colorado Department of Health
codifies occupation using The Standard Occupational Classification Manual (U.S. Department of
Commerce, Office of Federal Statistical Policy and Standards, 1980). They codify level of education
using The Data Management Section Statistical Abstract File Description. 1970 Through the Current
Year (The Colorado Department of Health, 1989). For a total of thirty-seven cases, eight (11% of the
total population) fell into Class Level II, fifteen (20% of the total population) into Class Level m, and
fourteen (19% of the total population) into Class Level IV. There was no incidence of Class Level I
or V. When analyzed by race and origin, Class Level n had seven White/Nonhispanic women and
one Black/Nonhispanic woman. Class Level in had seven White/Nonhispanic women, five
Black/Nonhispanic women, one Black/Other and Unknown Spanish woman, one White/Other and
Unknown Spanish woman, and one White/Mexican woman. Class Level IV had six
Black/Nonhispanic women, six White/Nonhispanic women, one White/Other and Unknown Spanish
woman, and one Black/Other and Unknown Spanish woman (see Table 4.9).
YEARS OF EDUCATION BY RACE AND ORIGIN
Race/ YEARS OF EDUCATION
0-6 Years 7-9 Years 10 Years 11 Years SUBTOTAL
# % # % # % # % # %
W/Nh 2 3% 1 1% 2 3% 3 4% 8 11%
W/OS 2 3% 2 3% 0 0% 1 1% 5 7%
W/M 0 0% 0 0% 0 0% 0 0% 0 0%
B/Nh 1 1% 1 1% 2 3% 2 3% 6 8%
B/OS 0 0% 0 0% 1 1% 0 0% 1 1%
Total 5 7% 4 5% 5 7% 6 8% 20 27%
Legend: W/Nh= White/Nonhispanic W/OS= White/Other & Unknown Spanish %=Percent of 75
W/M= White/Mexican B/Nh= Black/Nonhispanic B/OS=Black/Other & Unknown Spanish
Table 4.8 YEARS OF EDUCATION BY RACE AND ORIGIN (Continued)
Race YEARS OF EDUCATION
12 Years 13-15 Years 16 + Years SUBTOTAL TOTAL MEAN
# % # % # % # %
W/Nh 13 17% 6 8% 7 9% 26 35% 34 45% 8.5
W/OS 1 1% 0 0% 0 0% 1 1% 6 8% 10
W/M 2 3% 0 0% 0 0% 2 3% 2 3% 12
B/Nh 5 7% 3 4% 2 3% 10 13% 16 21% 12.5
B/OS 3 4% 0 0% 0 0% 3 4% 4 5% 11
Total 24 32% 9 12% 9 12% 42 56% 62 83%
Table 4.9 SOCIOECONOMIC STATUS: Utilizing the Hollingshead Two Factor
Index of Social Class
Race/ CLASS LEVEL
I II m IV V TOTAL
# % # % # % # % # % # %
W/Nh 0 0% 7 9% 7 9% 6 8% 0 0% 20 27%
W/OS 0 0% 0 0% 1 1% 1 1% 0 0% 2 3%
W/M 0 0% 0 0% 1 1% 0 0% 0 0% 1 1%
B/Nh 0 0% 1 1% 5 7% 6 8% 0 0% 12 16%
B/OS 0 0% 0 0% 1 1% 1 1% 0 0% 2 3%
Total 0 0% 8 11% 15 20% 14 19% 0 0% 37 49%
Legend: W/Nh=White/Nonhispanic W/OS=White/Other & Unknown Spanish
W/M=White/Mexican B/Nh=Black/Nonhispanic B/OS=Black/Other & Un-
%=Percent of 75 known Spanish
Occupation and Industry
The Colorado Department of Health codes occupation and industry using The Standard
Occupational Classification Manual (U.S. Department of Commerce, Office of Federal Statistical
Policy and Standards, 1980) and The Standard Industrial Classification Manual (Executive Office of
the President, Office of Management and Budget, 1987), respectively. There were a total of forty-six
women for whom the variable of occupation could be analyzed (see Table 4.10). Of these, the
occupation most cited was Registered Nurse, with five incidents (7% of the total population). The
occupations Waitress, Technician, and Computer Operator each showed two incidents (3%
each of the total population). The title Construction Laborer was cited three times (4% of the total
population) and Manager, Properties and Real Estate was cited four times (5% of the total
population). Interestingly, of the forty-seven cases for which the variable could be analyzed, the
industry title Hospital was cited most often (see Table 4.11), with a total of eleven incidents (15%
of the total population). The industries Furniture and Home Furnishing Store, Miscellaneous
Retail Store, Personnel Supply Service, and Social Services had each two cases (3% each of the
total population). The industry Construction was cited four times (5% of the total population) and
Eating and Drinking Place was mentioned five times (7% of the total population). Of the
remaining records, each was cited only once.
Manner of Death
The Colorado Department of Health codes the variable of manner of death using The Data
Management Section Statistical Abstract File Description. 1970 Through the Current Year (The
Colorado Department of Health, 1989). For seventy-five cases, fifty-nine (79%) were recorded as a
natural death, one (1%) as a suicide, and fourteen (19%) as an unknown cause of death. When
combined with race and origin, the Unknown category included one White/Irish woman (1% of the
total population), one Black/Other Southern European (1% of the total population), and twelve
White/American women (16% of the total population). The Suicide category included one
White/Nonhispanic woman (1% of the total population). Finally, the Natural Death category
included thirty-five White/Nonhispanic women (47% of the total population), sixteen
Black/Nonhispanic women (21% of the total population), four White/Other and Unknown Spanish
women (5% of the total population), four Black/Other and Unknown Spanish women (5% of the total
population, and two (3% of the total population) White/Mexican women (see Table 4.12).
Occupation NUMBER OF PERCENT OF
Administrative Support 1 1%
Billing Clerk 1 1%
Bookkeeper, Accountant 1 1%
Cashier 1 1%
Clinical Lab. Technologist 1 1%
Computer Operator 2 3%
Construction Laborer 3 4%
Cook 1 1%
Designer 1 1%
Electronic Technician 1 1%
General Office Clerk 1 1%
Housekeeper 1 1%
Laborer (not Construction) 1 1%
Licensed Practical Nurse 1 1%
Maid 1 1%
Manager, Administrator 1 1%
Manager, Prop. & Real Estate 4 5%
Nursg Aid,Orderly, Attendant 1 1%
Optical Goods Worker 1 1%
Machine Operator 1 1%
Painter, Constn., Maintenance 1 1%
Mail Superintendent 1 1%
Production Inspector 1 1%
Real Estate Sales 1 1%
Registered Nurse 5 7%
Secretary 1 1%
Sorter (excpt. agriculture) 1 1%
Supervisor, Sales 1 1%
Supervisor, Food Preparation 1 1%
Supervisor, Production 1 1%
Teacher 1 1%
Teacher, Secondary 1 1%
Technician 2 3%
Waitress 2 3%
TOTAL 46 61%
INDUSTRY NUMBER OF WOMEN PERCENT OF WOMEN
Agricultural Product., Livestock 1 1%
Banking 1 1%
Beverage Industry 1 1%
Construction 4 5%
Direct Selling Establishment 1 1%
Eating and Drinking Place 5 7%
Educational Service 1 1%
Electronic Computing Equip. 1 1%
Elementary & Secondary Schools 1 1%
Fumiture/Home Furnish. Store 2 3%
Hospital 11 15%
Hotel and Motel 1 1%
Industry 1 1%
Insurance 1 1%
Legal Service 1 1%
Lodging (except hotels & motels) 1 1%
Misc. Entertainment Svcs. 1 1%
Misc. Personal Services 1 1%
Misc. Retail Store 2 3%
Nursing & Personal Care Facilit. 1 1%
Personnel Supply Office 2 3%
Real Estate 1 1%
Social Services 2 3%
Toys, Amusement, Sporting Gds. 1 1%
U.S. Postal Service 1 1%
Warehouse and Storage 1 1%
TOTAL 47 63%
fflV/AIDS Incidence Per Year. 1987-1992
The Colorado Department of Health codifies the year using The Data Management Section
Statistical Abstract File Description. 1970 Through the Current Year (The Colorado Department of
Health, 1989). A total of seventy-five cases had the year recorded In 1987, eight (11% of the total
MANNER OF DEATH BY RACE AND ORIGIN
Race/ Unknown Suicide Natural TOTAL
Origin Number Percent Number Percent Number Percent Number Percent
W/Nh 0 0% 1 1% 35 47% 36 48%
W/A 12 16% 0 0% 0 0% 12 16%
w/os 0 0% 0 0% 4 5% 4 5%
W/M 0 0% 0 0% 2 3% 2 3%
W/I 1 1% 0 0% 0 0% 1 1%
B/Nh 0 0% 0 0% 16 21% 16 21%
B/OS 0 0% 0 0% 4 5% 4 5%
B/OSE 1 1% 0 0% 0 0% 1 1%
TOTAL 14 19% 1 1% 61 81% 75 100%
Legend: W/Nh=White/Nonhispanic W/A=White/American W/OS=White/Other & Un- W/M=White/Mexican W/I=White/Irish known Spanish B/Nh=Black/Nonhispanic B/OS=Black/Other & Unknown Spanish B/OSE=Black/Other Southern European (excludes Spain)
population) women died with HTV/AIDS listed as the primaiy, secondary, or underlying cause or with
HIV/AIDS listed as an aggravating condition at the time of death. In 1988 there were six cases (8%
of the total population), in 1989 there were six (8% of the total population), in 1990 there were
sixteen (21% of the total population), in 1991 there were eighteen (24% of the total population), and
in 1992 there were twenty-two women (29% of the total population) who died from HTV/AIDS
complications. When combined with race and origin, in 1987, eight White/American women died
(11% of the total population). In 1988, one White/Irish woman died (1% of the total population), one
Black/Other Southern European (excludes Spain) died (1% of the total population), and four
White/American women died (5% of the total population). In 1989, one White/Nonhispanic woman
died (1% of the total population), two Black/Other and Unknown Spanish women died (3% of the
total population), and three Black/Nonhispanic women died (4% of the total population). In 1990,
three Black/Nonhispanic women (4% of the total population) and thirteen White/Nonhispanic women
died (17% of the total population). In 1991, three Black/Other and Unknown Spanish women (4% of
the total population), six Black/Nonhispanic women (8% of the total population), and nine
White/Nonhispanic women died (12% of the total population). Finally, in 1992, two White/Mexican
women (3% of the total population), four White/Other and Unknown Spanish women (5% of the total
population), four Black/Nonhispanic women (5% of the total population), and twelve
White/Nonhispanic women (16% of the total population) died from HIV/AIDS related complications
(see Table 4.13).
Table 4.13 HIV/AIDS INCIDENCE PER YEAR, 1987-1992
BY RACE AND ORIGIN
1987 1988 1989 SUBTOTAL
Race/Ori Number Percent Number Percent Number Percent Number Percent
gin W/Nh 0 0% 0 0% 1 1% 1 1%
W/A 8 11% 4 5% 0 0% 12 16%
w/os 0 0% 0 0% 0 0% 0 0%
W/M 0 0% 0 0% 0 0% 0 0%
W/I 0 0% 1 1% 0 0% 1 1%
B/Nh 0 0% 0 0% 3 4% 3 4%
B/OS 0 0% 0 0% 2 3% 2 3%
B/OSE 0 0% 1 1% 0 0% 1 1%
TOTAL 8 11% 6 8% 5 7% 20 27%
Legend: W/Nh=White/Nonhispanic W/A=White/American W/M=White/Mexican
W/I=White/Irish W/OS=White/Other & Unknown Spanish B/Nh-Black/Nonhispanic
B/OS=Black/Other & Unknown Spanish B/OSE=BIack/Other Southern European (excludes
HIV/AIDS INCIDENCE PER YEAR, 1987-1992
BY RACE AND ORIGIN (Continued)
1990 1991 1992 SUBTOTAL TOTAL
Race/ Num Per Num Per Num Per Num Per Num Per
Ori ber cent ber cent ber cent ber cent ber cent
gin W/Nh 13 17% 9 12% 12 16% 34 45% 35 47%
W/A 0 0% 0 0% 0 0% 0 0% 12 16%
w/os 0 0% 0 0% 4 5% 4 5% 4 5%
W/M 0 0% 0 0% 2 3% 2 3% 2 3%
W/I 0 0% 0 0% 0 0% 0 0% 1 1%
B/Nh 3 4% 6 8% 4 5% 13 17% 16 21%
B/OS 0 0% 3 4% 0 0% 2 3% 4 5%
B/OSE 0 0% 0 0% 0 0% 0 0% 1 1%
Total 16 21% 18 24% 22 29% 55 73% 75 100
Common Illnesses in the Population
The Colorado Department of Health codes disease utilizing The International Classification
of Diseases. 9th Revision. 4th Edition (U.S. Department of Health and Human Services, 1991). A
total of sixty-five disease conditions were recorded in the data. The total number of disease incidence
was two hundred thirty-seven (see Table 4.14). Of course, the disease range code recorded most often
was Infectious and Parasitic Diseases, which showed one hundred sixty-four times in the data
(72.2% of all diseases mentioned). The code ranges Injury and Poisoning and Diseases of the
Nervous System and Sense Organs showed eleven cases each (4.8% of all diseases mentioned). The
code ranges Symptoms, Signs and Ill-Defined Conditions, Diseases of the Digestive System,
Diseases of the Circulatory System, and Endocrine, Nutritional and Metabolic Diseases and
Immunity Disorders was cited eight times (3.5% of all diseases mentioned). Disease of the
Respiratory System was mentioned six times (2.6% of all diseases mentioned), Disease of the Blood
and Blood Forming Organs was mentioned five times (2.2% of all diseases mentioned),
Neoplasms (tumors) was cited three times (1.3% of all diseases mentioned), Mental Disorders
was cited twice (.9% of all diseases mentioned), and Congenital Anomalies (anomalies dating from
birth) was mentioned once (.4% of all diseases mentioned).
NUMBER AND PERCENT OF INCIDENTS BY DISEASE
NAME AND RANGE CODES
Disease Range Name Disease Range Code Number Number of Incidents Percent of Incidents
Infectious & Parasitic 001-139 164 69%
Neoplasms 140-239 3 1.3%
Endocrine, Nutritional, & Metabolic Disease & Immunity Disorders 240-279 8 3.4%
Disease of Blood & Blood Forming Organs 280-289 5 2.1%
Mental Disorders 290-319 2 .8%
Disease of Nervous Sys. & Sense Organs 320-389 11 4.6%
Disease of Circulatory Sys. 390-459 8 3.4%
Disease of Respiratory Sys. 460-519 6 2.5%
Disease of Digestive Sys. 520-579 8 3.4%
Disease of Genitourinary Sys 580-629 0 0%
Complications of Preg., Ch.Birth, & the Puerperium 630-676 0 0%
Disease of Skin & Subcu- taneous Tissue 680-709 0 0%
Disease of Musculoskeletal Sys. & Connective Tissue 710-739 2 .8%
Congenital Anomalies 740-759 1 .4%
Certain Conditions Orig. in Pre-natal period 760-779 0 0%
Symps, Signs, & Ill-Def Condits. 780-799 8 3.4%
Injury & Poisoning 800-999 11 4.6%
Using death certificates to analyze the demographic and socioeconomic background variables
of women in Colorado who have died of HIV/AIDS has been very helpful and revealing. Even
though some of the results mirrored the existing findings, others were quite unexpected. For
instance, all of the national HIV literature to date has reported the highest incidence of female HIV
infection during the child bearing years, ages fifteen through forty-five. According to the Kinsey
Reports, womens sexual activity increases from ages fifteen through thirty, then plateaus from age
thirty through fifty (Katchadourian, 1985). As heterosexual sex is a large behavioral risk for women,
and considering a diagnosis to death period, a great many of the women who have died from AIDS
should be toward the middle to end and post child bearing years. Furthermore, since another high
risk category for women is intravenous drug use and since one would expect that people who abuse
drugs would have a lesser life expectancy than the norm, these HTV infected women, too, should
meet death at about the same age range. The data analysis for the variable of age concurred with the
literature in that the overall median age of the decedents was 69.5. Furthermore, the disparity in
analysis of age by race showed a 6.5 year differential, in favor of White women, which is also
consistent with the literature. While White women had a median age of 57 years, Black women had a
median age of only 50.5. This may be an indication of the different power, status, and class
differentials between Black and White women. When race and origin were combined with age in the
analysis, most of the results concurred with the literature,as well. First, White/Nonhispanic,
White/American, and White/Irish women tended to live longer than Black/Nonhispanic, Black/Other
and Unknown Spanish, and Black/Other Southern European (excludes Spain). Continuing in this
vein, Black/Nonhispanic, Black/Other and Unknown Spanish, and Black/Other Southern European
(excludes Spain) women tended to live longer than the White/Mexican and White/Other and
Unknown Spanish women.
Another result which concurred with the literature was that of the variable of residence. The
largest concentrations of HIV/AIDS infection are found in urban, metropolitan areas, such as
New York, San Francisco, and Miami (Chu and Diaz, 1993; Dicus, 1994). The present study
demonstrated, overwhelmingly, that this is so, with Denver and El Paso Counties showing the
greatest number of woman residents (69% of the total population). Also, disparity was evident
between Black and White women, showing that White women were more likely to have resided in a
suburb than Black women. This also may be an indication of the differential class and income levels
of the two groups. Unfortunately, there were not enough cases to determine if this trend continued
between women of White/Hispanic race and origin and White/Nonhispanic/American women.
A third variable which mirrored the literature was that of socioeconomic status. The
literature which has been cited to date indicates that socioeconomic status is, indeed,
directly related to illness. Increased incidence of illness and decreased access to health care have
been shown to be definitive indicators of low socioeconomic status. In this study, 11% of the women
fell into Class Level II, 20% of the women fell into Class Level HI, and 19% fell into Class Level IV.
Class Level n is approximately the equivalent of the middle class, Class Level III is approximately
equal to the lower-middle class, and Class Level IV is the approximate equivalent of the working
class. Although no women were classified as Class Level V, the data show a definite lower class
trend. Furthermore, it is important to note that none of the women were classified as Class Level I,
A fourth variable which is strongly reflected in the literature, and confirmed by this study, is
that of HIV/AIDS incidence per year. The amount of multiplication is quite severe, as this study
starts out with eight incidents in 1987, yet ends with 22 incidents in 1992. As the Centers for Disease
Control has updated its definitions of HIV/AIDS, more women are being identified and diagnosed
correctly. This fact, coupled with an expected increase in incidence itself, would lead one to believe
that the trend of increase will probably continue for years to come.
As stated previously, while many of these results concurred with the literature, some did not.
One variable which was not anticipated in the literature was that of race, ethnicity, and origin.
Surprisingly, the majority of the women in this study were White, rather than Black. When race was
combined with origin, the number of minorities did increase but still showed more White women.
White/Nonhispanic/American/Irish women totaled approximately 60% of the population, whereas
White/Other and Unknown Spanish/Mexican and Black/Nonhispanic/Other and Unknown
Spanish/Other Southern European (excludes Spain) totaled approximately 40% of the population.
This disparity may be, in part, the result of Colorados low Black-to- White ratio. However, Colorado
does have a substantial Hispanic population and these people are not reflected here. It is very
interesting to note that, while there were forty-three White/Nonhispanics and twelve
White/Americans, there were eight Black/Nonhispanics and no Black/Americans. Could it be that
doctors are less likely to see Black people as Americans? Or were none of these Black people
United States citizens?
The variable of marital status was another which contradicted the literature in some respects.
The Centers for Disease Control (CDC) performed a demographic survey of two hundred ninety-one
women with HTV/AIDS from January 1,1990, through July 31, 1992 (CDC, 1993). Forty-five
percent of these women had never been married, fourteen percent were currently married, twenty-six
percent were separated or divorced, and seven percent were living with a partner. The results of the
present study showed results almost exactly even with the CDC study in the indicator of Separated or
Divorced, with 29% showing divorce status. However, only 21% of the cases in the present study
were recorded as Single, whereas 45% were recorded as single in the CDC study. Also, in this
study 39% of the women were married, and only 14% were in the CDC study. The married
indicator is particularly intriguing and leads to many questions. For instance, were these women
infected with HIV before they married, or did they contract it during their marriage? Also, if they
contracted HTV during their marriage, was it from their spouse or extramarital intercourse? Or, was
the transmission of the virus related to intercourse at all? The CDC study did not compute the
Widowed indicator, so there exists nothing from them with which to compare the present results.
Additionally, while the CDC study measured women classified as Living with a Partner, the
Colorado Department of Health had no indicator for this variable.
Another variable which contradicted the literature in some respects was that of education.
The level of education of these women was a little higher than reported in the literature. By far, for
Black and White women, the majority had a high school level education and some even had higher
educations. It is reasonable to assume that, since women hold a low socioeconomic class level in the
United States, their education would be lacking. Instead, what was revealed was an almost bell curve-
like distribution of levels of education, a smaller percentage at the highest and lowest ends, with the
majority in the middle. Still, it is important to remember that a high school level education does not
offer the prestige and status that a college level.education would in this country.
Another way in which the results of education analysis contradicted the literature was in the fact that
the mean level of education for White women was lesser than that of Black women, coming to 8.5
and 12.5, respectively. Also, the mean levels of education for White/Mexican and White/Other and
Unknown Spanish women were larger than that of White/Nonhispanic. White/Mexican women had a
mean level of education equaling 12 years and White/Other and Unknown Spanish women had a
mean level of education equaling 10 years. On the other hand, White/Nonhispanic women had a
mean level of education equaling only 8.5. However, one way in which this variable does reflect the
literature is in years of post-secondary schooling. Thirteen White women received post-secondary
schooling, whereas only five Black women did
Analysis of the occupation/industty variables revealed results which were inconsistent with
the literature, as well. It was believed that the most common occupations/industries would be of the
low prestige, low paying, feminized type. While many occupations/industries named were of the
feminized type, the occupation most cited was Registered Nurse which is not necessarily low pay or
prestige. The most cited industry was Hospital and citations for this industry were more than the
number of registered nurses, so some of the Technician occupations fell into Hospital classification as
well. Regarding the other occupations recorded Computer Operator, Construction Laborer,
Technician, and Waitress were, in fact, more of what has been reported before, in so far as they are
low pay, prestige, and status. However, a significant number of women occupied the Manager,
Properties and Real Estate occupation, which is a more prestigious position. The remaining industry
classifications mirrored the trend of occupations. Eating and Drinking Place and Construction were
recorded often and Furniture and Home Furnishing Store, Miscellaneous Retail Store, Personnel
Supply Office, and Social Services had two incidents each. All of these industries are on the
moderate to low end of notable industries.
The manner of death variable was not noted, specifically, in the literature. One would
assume, though, that most deaths would be natural. This was shown in the present study to be true,
with 81% of incidents being of natural death. One woman commited suicide and fourteen women
died of an unknown manner of death. This variable spurs many questions. For instance, the incident
of suicide, although small and seemingly insignificant, leads to questions of whether there is a
possible trend of suicide among HIV positive women. With regard to the women who died in an
unknown manner, most of these occured in the early years of the study, 1987-1989. It is possible that
this variable was not identified on death certificates before 1990.
Common illnesses studied in the data were surprising, not so much because of what they
showed, but what they did not show. There was not one mention of Disease of the Genitourinary
System or Complications of Pregnancy, Childbirth, and the Puerperium, meaning that the
gynecological manifestations cited in the literature were not present in this population..
Obviously, the HIV/AIDS diagnosis made the category of Infectious and Parasitic Disease increase, as
a full 69% of all 237 diagnoses fell under this heading. The other classifications, Diseases of the
Nervous, Circulatory, Respiratory, and Digestive Systems are expected in any HIV/AIDS patient, and
so shed little light on this variable.
Limitations. Justifications, and Conclusions
No research strategy is perfect and every study has its flaws. And, while it is firmly believed
that this strategy is presently the best and most efficient way to describe the women in Colorado with
HIV/AIDS, it is no exception. One of the biggest quandaries of this study is whether the study of
women who have died from HIV/AIDS will tell us anything about the women who are presently
living with HIV/AIDS. For instance, recently some doctors have been commenting on the
extraordinary life spans of some of their HTV patients. No formal research exists, to date, on this
phenomena. Yet, suppose it is true and that there are women whose disease progression is very slow.
These women would not be fully represented in this study. Also, women who never were diagnosed
correctly will not be represented here. A second limitation is that the variables of
race/ethnicity/origin were not very thoroughly or accurately defined by the Colorado Department of
Health. Another limitation of this study is the assumption that people will be diagnosed honestly and
correctly on their death certificates. It is possible that a well meaning doctor could deliberately
substitute a slightly off-target diagnosis on the death certificate, perhaps to save the family members
from shock or embarrassment. An additional limitation is in relation to the Hollingshead Two Factor
Index of Social Class. This calculation can be somewhat unstable with regard to levels of education
easily skewing the outcome. Initially, this project included the use of the Duncan Socioeconomic
Index. However, this calculation needs an amount of income to be complete and trustworthy. Since
income was not included on the death certificates and because no reliable average income index exists
for the occupation/industrial classifications used here, it was decided that Hollingsheads Index would
be most useful.
Despite these shortcomings, though, this study offers a complete and thorough way to draw a
picture of the women in Colorado who have died from HTV/AIDS complications. Analysis of
variables such as socioeconomic status, occupation, industry, and marital status make this study, and
the use of secondary data vital statistics, a valuable contribution to the basis of knowledge already
existing on this particular population. Needless to say, a lot of research is still needed in this area.
For instance, a survey of women with HIV which analyzes the same variables as the present one, but
includes the contextual information which this study lacks, is called for. With regard to methodology,
there is a dearth of experimental research regarding the psychosocial issues of women with
HTV/AIDS. These and many other research strategies are being considered for the future. Today,
women are at a dangerously high risk level for HTV/AIDS infection and the more research performed
in this area will be all that much more helpful and preventive for them.
LETTER OF PROTOCOL TO
THE COLORADO DEPARTMENT OF HEALTH
The Colorado Department of Health
Health Statistics Section
4300 Cherry Creek Drive South
Denver, Colorado 80222-1530
Dear Ms. Garrett,
This is a letter of introduction, inquiry, and request. My name is Katherine Lineberger and I
am presently undertaking graduate work in Sociology at the University of Colorado at Denver. The
focus of my research and masters thesis is issues of women and HTV/AIDS. My studies of the
existing literature and research in this area have led me to determine that there are large holes in the
demographic descriptions of this population. It has occurred to me that, perhaps, a study of the death
certificates of women who have died of AIDS related complications or opportunistic infections related
to HTV disease would glean a more accurate picture of these women than we now have.
Presently, I am designing a research proposal to perform just this style of task. My inquiries
at the Department of Health led me to Katie Meng in your statistics group. She informed me that
Colorado keeps death certificates as confidential and that my proposal would need to be approved,
through you, before I could begin work. Further discussion with Ms. Meng revealed that it would be
possible for the Department of Health to provide aggregate data on my population, without
compromising the confidentiality of the individuals death certificates. I would very much like to take
advantage of this latter opportunity, with your permission. Ms. Meng also said that there would be
confidentiality papers to be signed before commencement of research. Per your advice and
requirement in this area, I would fully cooperate with anything you might need from me.
Enclosed is a copy of my proposal, to assist you in understanding the nature and protocol of
my study. If you have any questions at all, please do not hesitate to call me anytime at (303) 936-
6630. Thank you very much for your time and attention to this matter. I hope to hear from you soon.
SAMPLE COLORADO DEATH CERTIFICATE
AOKS-1S 1-s 1-41)
The ACT UP / New York Women and AIDS Book Group
1990 Women, AIDS, and Activism. Boston: South End Press.
1990-91 Black Women and HTV/AIDS. Siecus Report (December/Januaiy): 8-10.
Barnett, Carol Beth
1993 The Forgotten and Neglected: Pregnant Women and Women of Childbearing
Age in the Context of the AIDS Epidemic. Golden Gate University Law
Review 23: 863-898.
Benson, Denice J.D., and Catherine Maier
1990 Challenges Facing Women with HIV. Focus: A Guide to AIDS Research and
Counseling 6, 1 (December): 357-358.
Bosk, Charles L., and Joel E. Frader
1994 AIDS and Its Impact on Medical Work: The Culture and Politics of the Shop
Floor. pp. 221-233 in Peter Conrad and Rochelle Kern (eds.) The Sociology of
Health and Illness: Critical Perspectives. New York: St. Martins Press.
1992 Pollutants, Criminals, and Incubators: The Conceptualization of Women Under
State HIV/AIDS Law 1983 to 1991. IRIS (Spring/Summer): 11-20.
Brown, Sharon C.
1993 AIDS in Women and Children. U.S. Pharmacist: Womens Health Issues
Buehler, James W., Hanson, Debra, and Susan Y. Chu
1992 The Reporting of HTV/AIDS Deaths in Women. American Journal of Public
Health 82,11 (November): 1500-1504.
Cambridge, Barbara S.
1988 Women and HIV Disease. pp. 281-293 in Ted Eidson (ed.) AIDS Caregivers
Handbook. New York: St. Martins Press.
Campbell, Carol A.
1990 Women and AIDS. Social Science and Medicine 30,4: 407-415.
Chu, Susan Y., and Theresa Diaz
1993 Living Situation of Women with AIDS. Journal of Acquired Immune
Deficiency Syndromes 6, 4: 431-432.
Congress of the United States
1992 The CDCs Case Definition of AIDS: Implications of Proposed Revisions.
Washington, D.C.: Office of Technology Assessment.
1992 The Story of Women and AIDS: The Invisible Epidemic. New York: Harper
1994 Phone Conversations.
Ehrenreich, Barbara, and Deirdre English
1973 Complaints and Disorders: The Sexual Politics of Sickness. New York: The
Farmer, Paul, Lindenbaum, Shirley, and Mary-Jo Delvecchio Good
1992 Women and AIDS. Culture, Medicine and Psychiatry : 386-397.
Gayle, Jacob A., Selik, Richard M., and Susan Y. Chu
1991 Surveillance for AIDS and HIV Infection Among Black and Hispanic
Children and Women of Childbearing Age, 1981-1989. Sixth Annual
Rocky Mountain Conference on AIDS (January): 344-351.
Graber, Mark A.
1994 The Spectrum of HTV Disease in Women. Physician Assistant (January):
1990 Missing Persons: African American Women, AIDS and the History of
Disease. Radical America 24, 2: 7-23.
Hankins, Catherine A., and Margaret A. Handley
1992 HIV Disease and AIDS in Women: Current Knowledge and a Research
Agenda. Journal of Acquired Immune Deficiency Syndromes 5 : 957-
Harris Mullan, Kathleen
1993 Work and Welfare Among Single Mothers in Poverty. American Journal
of Sociology 99,317-352.
Katchadourian, Herant A.
1985 Fundamentals of Human Sexuality, Fourth Edition. New York: Holt,
Rinehart and Winston.
Kitson, Gay C.
1992 Portrait of Divorce: Adjustment to Marital Breakdown. New York: Guilford
1987 AIDS: The Ultimate Challenge. New York: Macmillan Publishing Company.
1994 Phone Conversation. March 28.
Leonard-Syme, S., and Lisa F. Berkman
1994 Social Class, Susceptibility, and Sickness. pp. 29-35 in Peter Comad
and Rochelle Kem (eds.), The Sociology of Health and Illness: Critical
Perspectives. New York: St. Martins Press.
1990-1991 Latinas and HTV/AIDS. Siecus Report (December-January): 11-17.
1977 Handbook of Research Design and Social Measurement, Third Edition.
New York: David McKay Company, Incorporated.
Murphy, Julien S.
1988 Women with AIDS: Sexual Ethics in an Epidemic. pp. 65-77 in Inge B.
Corless and Mary Pittman-Lindeman (eds.), AIDS: Principles, Practices
and Politics. Washington: Hemisphere Publishing Corporation.
1993 Dangerous to Your Health: Capitalism in Health Care. New York:
Monthly Review Press.
Ollenburger, Jane C., and Helen A. Moore
1992 A Sociology of Women: The Intersection of Patriarchy, Capitalism and
Colonization. Englewood Cliffs, New Jersey: Prentice Hall.
1992 Society Facing AIDS. Current Sociology 40, 3 (Winter): 11-23.
1992 AIDS: A Problem for Sociological Research. Current Sociology 40, 3
1990 A Global View of Women and HTV. Focus: A Guide to AIDS Research
and Counseling 6,1 (December): 359.
1991 The Social Construction of AIDS, Heterosexism, Racism, and Misogyny:
And the Challenges Facing Women of Colour. Resources for Feminist
Research 20, 3-4 (Fall-Winter): 115-123.
U.S. Bureau of the Census
1960 Methodology and Scores of Socioeconomic Status. Washington D.C.:
U.S. Department of Commerce.
U.S. Department of Health and Human Services
1991 The International Classification of Diseases: Ninth Revision, Clinical
Modification. Fourth Edition. Washington, D.C.
Weiss, Gregory L., and Lynne E. Lonnquist
1994 The Sociology of Health, Healing, and Illness. New Jersey: Prentice Hall.
Wiener, Lori S.
1991 Women and Human Immunodeficiency Virus: A Historical and Personal
Psychosocial Perspective. Social Work 36, 5: 375-378.
Withers-Osmond, Marie, Wambach, K.G., Harrison, Dianne, Byers, Joseph, Levine, Phillipa,
Imershein, Allen, and David M. Quadagno
1993 The Multiple Jeopardy of Race, Class, and Gender for AIDS Risk Among
Women. Gender and Society 7,1 (March): 99-120.
Zarembka, Arlene, and Katherine M. Franke
1990 Women in the AIDS Epidemic: A Portrait of Unmet Needs. Saint Louis
University Public Law Review 9: 519-541.
Zola, Irving Kenneth
1994 Medicine as an Institution of Social Control. pp. 392-402 in Peter Conrad
and Rochelle Kern (eds.) The Sociology of Health and Illness: Critical
Perspectives. New York: St. Martins Press.